﻿* Encoding: UTF-8.
* Encoding: .

****NOTE that syntax up to R&R designation was syntax for original NMS submission.

**** Data only includes those who uploaded data, had more 2000 site visits and 7 days of data. 


*** Media exposure variables.
* leftlgw1 "Left-leaning site visit count log-W1".  

* ritelgw1 "Right-leaning site visit count log-W1".  

* cntrlgw1 "Center site visit count log-W1".  

* fnlgw1 "Fake news site visit count log-total-W1".

*********

***Demographics.

*Party ID in wave 1 = variable named 'party1'

* polIntw1 'political interest w1'.

* genderw1 'Gender-women coded high-w1'.

* agew1 'Age-W1'.

* eduw1 'Education-W1'.

* racew1 'Race-White coded high-W1'.

******
control variables.

* usersw1 'Other users of browser-yes coded high-w1'.

* mtldevw1 'Respondent uses other devices.'

*******

*False Beliefs. Higher values indicate greater accuracy.

***Claim 1. Putin Ordered interference in US Election.
*Reverse coded to show that responses indicating that its false are inaccurate.

* Putinw1 'Putin ordered US Election Interference-w1'.


***Claim 2. Parkland school shooter was a member of white supremacist group. 

* Prklndw1 'Parkland school shooter is white supremacist-w1'.

*** Claim 3. Hillary Clinton lied to the FBI about her email server.
 
* HCw1 'Clintion lied to FBI about email server-w1'.


***Claim 4. More Gun owners reduces gun crime. 

* Gunw1 'More gun owners leads to less crime-w1'

*** Claim 5. Meuller cleared Trump of wrongdoing, finding no collusion with Russia.

* Muellw1 'Mueller cleared trump in Russia probe-W1'.


*** Claim 6. Trump knew about attempts by his campaign to work with Russia.

* DTw1 'Trump knew about campaign help from Russia-w1'.

***********
Analysis.

***Regression analyses for initial submission.


REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Putinw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Prklndw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT HCw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Gunw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Muellw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT DTw1 
  /METHOD=ENTER leftlgw1 ritelgw1 cntrlgw1 fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.


************************
************************
************************

*R&R. In the reivison we excluded 4 polling sites bbased on a request from a reviewer. We recoded the exposure variables below without those sites and reran all analyses. These are the analyses in the final version of the paper.*
*(pewreserach.com, publicpolicypolling.com, poll.qu.edu, and gallup.com excluded).

* Eady Scores--no polls.

* leftlgw1E "Left-leaning site visit count log-W1-Eady Scores".  

* ritelgw1E "Right-leaning site visit count log-W1-Eady Scores".  

* cntrlgw1E "Center site visit count log-W1-Eady Scores". 

**Regression with scores without polling sites.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Putinw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Prklndw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT HCw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.


REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Gunw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.

REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT Muellw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.


REGRESSION 
  /MISSING LISTWISE 
  /STATISTICS COEFF OUTS R ANOVA 
  /CRITERIA=PIN(.05) POUT(.10) 
  /NOORIGIN 
  /DEPENDENT DTw1 
  /METHOD=ENTER leftlgw1E ritelgw1E cntrlgw1E fnlgw1 party1 eduw1 racew1 polIntw1 usersw1 mtldevw1 agew1 
    genderw1.
